Skip to main content

Table 3 Micro-F1 scores for multi-label classification on PPI, Wikipedia, and CiteSeer datasets

From: SURREAL: Subgraph Robust Representation Learning

Algorithm PPI Wikipedia CiteSeer
  10% 50% 90% 10% 50% 90% 10% 50% 90%
DeepWalk 12.35 18.23 20.39 42.33 44.57 46.19 46.56 52.01 53.32
node2vec 16.19 20.64 21.75 44.38 48.37 48.85 50.92 52.49 56.72
Walklets 16.07 21.44 22.10 43.69 44.68 45.17 47.89 52.73 54.83
SURREAL 16.91 21.71 23.97 45.68 48.10 49.90 48.80 53.36 57.12
G.O. DWalk 36.85 19.08 17.55 7.90 7.91 8.03 4.80 2.59 7.13
G.O. N2vec 4.41 5.16 10.19 2.92 - 2.14 - 1.63 0.70
G.O. Walk 5.19 1.23 8.47 4.53 7.64 10.48 1.87 1.18 4.17
  1. Bolded numbers represent the best performance. By “G.O.” we denote “gain over”